Nickel-catalysed hydroalkoxylation reaction of 1,3-butadiene: Ligand controlled selectivity for the efficient and atom-economical synthesis of alkylbutenyl ethers

Article abstract of DOI:10.1002/chem.201300633

The nickel-catalysed hydroalkoxylation of butadiene is promoted by a nickel(0)/dppb catalyst (dppb=1,4-bis(diphenylphosphino)butane; see scheme). By following this new synthetic procedure, alkylbutenyl ethers are readily obtained from an alcohol and 1,3-butadiene with exclusion of dimerisation and telomerisation products. Copyright

Full text of DOI:10.1002/chem.201300633

COMMUNICATION
DOI: 10.1002/chem.201300633
Nickel-Catalysed Hydroalkoxylation Reaction of 1,3-Butadiene: Ligand
Controlled Selectivity for the Efficient and Atom-Economical Synthesis of
Alkylbutenyl Ethers
Sandra Bigot, Mohammed Samir Ibn El Alami, Alexis Mifleur, Yves Castanet,
Isabelle Suisse,* Andrꢀ Mortreux, and Mathieu Sauthier*[a]
Alkyl ethers are commonly obtained by the Williamson
reaction between an alcohol and a haloalkane in the pres-
ence of a base. In addition to often rather limited selectivity,
limitations arise from the formation of stoichiometric
amounts of salts and both more efficient and atom-economi-
cal processes are thus needed. The telomerisation reaction
allows the clean and efficient synthesis of octa-2,7-dienyl
ethers from 1,3-butadiene and an alcohol as nucleophile
(Scheme 1).[1] The use of palladium salts with phosphines,[2]
and more recently, carbene ligands[3] affords highly active
and robust catalysts thus allowing scale-up of the telomerisa-
tion reaction to industrial quantities for the selective synthe-
sis of 1-octene from 1,3-butadiene. In addition, this transfor-
mation has proven to be an efficient synthetic tool to graft
C8 alkadienyl chains onto glycerol, higher polyols, sugars or
starch for the synthesis of amphiphiles.[4] On the other hand,
the direct mono-addition of an alcohol to a diene leading to
shorter alkenyl ethers is much less studied. This transforma-
tion, also referred to as degenerative telomerisation or hy-
droalkoxylation, yields linear (cis and trans) 4 and branched
5 butenyl ethers (OC4; Scheme 1). However, this reaction is
rarely selective because the by-products that arise from telo-
merisation (linear 2 and branched 3 alkyl octadienyl ethers
(OC8)) and butadiene dimerisation[5] (1,3,7-octatriene 6 and
isomers) are also usually formed.[6]
Dewhirst first described the selective synthesis of butenyl
ethers 4 and 5 from butadiene and ethanol in the presence
of high loadings of rhodium trichloride as a catalyst.[7] With
p-allyl palladium chloride complexes and phenol as the sol-
vent, Smutny reported the formation of phenoxybutenes in
low yield (30%) along with ortho- and para-butenylphenols
and oligomerisation/dimerisation products.[8] Palladium-
based catalysts also allowed access to butenyl ethers with
high selectivities in dilute methanolic solutions.[9]
Reports of nickel-catalyzed telomerisation and hydro
oxyl tion reactions are rare. The use of a Ni(acac)2/NaBH4/
phenyldiisopropoxyphosphine (acac=acetylac tonate) mix-
ACHTUNGTRENNUNGalk-
A
E
E
N
ACHTUNGTRENNUNG
A
N
G
ACHTUNGTRENNUNG
ture had the propensity to produce a mixture of methoxybu-
tenes and methoxyoctadienes, along with some 1,3,7-octa-
trienes.[10] Attempts to obtain similar results with triphenyl-
phosphine, triphenylphosphite or triisopropoxyphosphine
were unsuccessful. Later work with different catalytic com-
binations of nickel sources, monophosphines and either phe-
nolic[11] or alcoholic nucleophiles gave similar results but
again without particular selectivity.[12–13] Selective and acces-
sible catalysts are necessary in order to generate butenyl
ethers and herein we report our recent progress in this field.
Our studies on the ethoxylation of butadiene were carried
out with different nickel precursors combined with simple
phosphines. In a typical experiment, the nickel precursor
Scheme 1. Nickel-catalysed transformations of 1,3-butadiene in the pres-
ence of an alcohol: Telomerisation, hydroalkoxylation and dimerisation
reactions.
[a] Dr. S. Bigot, Dr. M. S. I. El Alami, A. Mifleur, Prof. Y. Castanet,
Dr. I. Suisse, Prof. A. Mortreux, Prof. M. Sauthier
UCCS (Unitꢀ de Catalyse et de Chimie du Solide)
Universitꢀ Lille Nord de France - ENSCL
BP 90108, 59652 Villeneuve dꢁAscq (France)
Fax : (+33)3-20-33-63-01
bis(cyclooctadiene)nickel(0) or a mixture of bis(acetyl
ACHTUNGTRENNUNGacACHTUNGTRENNUNGe-
A
ACHTUNGTRENNUNG
were stirred for 5 min in ethanol in a Schlenk tube. Then,
liquid butadiene was added at À158C and the mixture was
heated to the desired temperature for 17 h. After cooling
E-mail: isabelle.suisse@ensc-lille.fr
mathieu.sauthier@univ-lille1.fr
Chem. Eur. J. 2013, 19, 9785 – 9788
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9785

Table 1. Hydroalkoxylation/telomerisation of butadiene with ethanol in
the presence of nickel catalysts.[a]
Entry Nickel source
Ligand
Butadiene
5
4
OC8 C8
Conv. [%] [%] [%] [%] [%]
1[b]
2
3
4
5
6
7
8
9
Ni
Ni
Ni
Ni
Ni
Ni
Ni
Ni
Ni
Ni
Ni
N
36
0
0
94
93
0
52
0
0
0
0
76
72
0
26
0
0
0
0
24
23
0
23
0
0
0
0
<2
<2
0
9
0
0
19
8
99
0
0
<2
5
0
42
0
0
23
5
dppb
dppb
Application of binap (2,2’-bis(diphenylphosphino)-1,1’-bi-
naphthyl) proved completely fruitless. Bis(aminophos-
phines) L1 and L2 were then synthesized in a single step
from diamines by phosphinylation with two equivalents of
chlorodiphenylphosphine in the presence of triethylamine.[14]
L1 and L2 have an electron-donor character due to the pres-
ence of two nitrogen atoms close to the phosphorous atoms,
which should have an effect on the selectivity, as found by
comparison of phosphine, phosphite and phosphonite li-
gands used for the nickel(0)-catalyzed addition of phenol to
butadiene.[11] Interestingly, both ligands show a reasonable
propensity to produce butenyl ethers (Table 1, entries 10
and 11). These results reinforce the importance of the ligand
backbone in inducing high hydroalkoxylation selectivity.
The effect of the dppb ligand on the observed selectivities is
particularly relevant. It seems clear that the flexibility of the
ligand backbone is a key factor in allowing the reaction to
proceed selectively towards C4 ethers. The dppb ligand has
previously resulted in unusual outcomes in several catalytic
reactions such as in the chemoselective palladium-catalyzed
methoxycarbonylation of 1,3-butadiene.[15] Unprecedented
reactivity[16] and linear regioselectivity have also been ob-
served in platinum-catalyzed hydroformylation of alkenes
by well-defined ligand bridged dinuclear platinum hydride
species.[17] Oligomeric (dppb)palladium–acyl complexes have
also been identified during studies of CO insertion into pal-
ladium alkyl bonds.[18] Accordingly, we suspect that such
0
86
100
0
50
65
0
8
22
10
11
[a] Conditions: butadiene=11.1 mmol, Ni/NaBH4/ligand/butadiene=
1:1:1.5:80, EtOH=10 mL, T=808C, t=17 h. acac=acetylacetonate,
cod=1,5-cyclooctadiene,
dppb=1,4-bis(diphenylphosphino)butane, dpppen=1,5-bis(diphenylphos-
phino)pentane, binap=2,2’-bis(diphenylphosphino)-1,1’-binaphthyl,
dppe=1,2-bis(diphenylphosphino)ethane,
dppp=1,3-bis(diphenylphosphino)propane. [b] 2 equivalents PPh3 with
respect to Ni.
and venting, the solution was analyzed by GC and the prod-
ucts were quantified with heptane as the internal standard.
Branched versus linear products could be differentiated. In
the presence of NiACHTUNGTRENNUNG(acac)2/NaBH4 and PPh3, as documented
previously, the reaction yields dimerization products with a
high selectivity (Table 1, entry 1).[5] On the other hand, the
use of the diphosphines dppe (1,2-bis(diphenylphosphino)-
ethane) and dppp (1,3-bis(diphenylphosphino)propane) did
not result in any butadiene conversion. Very surprisingly,
the use of the dppb (1,4-Bis(diphenylphosphino)butane)
ligand gave drastically different results (Table 1, entry 4),
yielding butenyl ethers 4 and 5 in higher than 95% selectivi-
ty with a 94% conversion of butadiene after 17 h at 808C.
The branched product 5 was obtained as the major butenyl
ether isomer (5/4=3). No reaction was observed if Ni
is used without sodium borohydride but very similar results
were obtained with [Ni(cod)2] (cod=1,5-cyclooctadiene;
Table 1, entries 4 and 5). A nickel(0) species thus appears to
be a key catalytic intermediate. The use of Ni(acac)2/NaBH4
proved to be very practical as compared with the use of air-
sensitive [Ni(cod)2]. This catalyst combination was thus re-
tained for the rest of the study. A rather limited conversion
of 1,3-butadiene is observed with the 1,5-diphenylphos-
phinopentane (dpppen; Table 1, entry 7) ligand, yielding a
quite large amount of dimerization products. In contrast to
the result obtained with the dppb ligand, the 5/4 ratio is
close to 1, thus showing that this regioselectivity can be
tuned according to the catalyst structure. These results also
suggest that the size of the bidentate ligand framework
plays a crucial role in the selectivity of the reaction. Xant-
phos shows no catalytic activity, possibly because of coordi-
nation of the oxygen atom of the ligand to the metal, thus
leading to highly coordinated complexes that exhibit low re-
activity. We then turned our attention towards diphosphines
that would give rise to a seven-atom heterocycle upon coor-
dination to a metal center.
G
oligoACTHGNUTERNNUmG eric (dppb)nickel allylic species could be responsible
for this particular reactivity/selectivity.
With dppb as a standard ligand, some reaction parameters
were then tuned in order to delineate the best reaction con-
ditions (Table 2). The amount of sodium borohydride was
G
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
Table 2. Optimisation of the hydroalkoxylation/telomerisation of buta-
diene with ethanol in the presence of the Ni
ACHTUGNTERN(NUNG acac)2/dppb/NaBH4 catalytic
system.[a]
G
ACHTUNGTRENNUNG
Entry
L/Ni
NaBH4/
Ni
Butadiene
Conv. [%]
5
[%]
4
[%]
OC8
[%]
C8
[%]
1
2
3
4
1.5
1.5
0.5
1
2.5
1.5
1.5
0.5
4
1
1
1
94
58
65
91
29
52
56
75
70
50
46
79
41
74
25
30
50
44
21
35
21
<2
<2
<2
5
<2
15
3
<2
<2
<2
5
<2
9
5
6[b]
7[c]
1
1
<2
[a] Conditions: butadiene=11.1 mmol, Ni/butadiene=1:80, EtOH=
10 mL, T=808C, t=17 h. [b] 1,3-butadiene=56 mmol, Ni/butadiene=
1:400, EtOH=10 mL, the reaction was performed in a steel high-pressure
reactor. [c] 1,3-butadiene=56 mmol, Ni/butadiene=1:400, EtOH=
20 mL, the reaction was performed in a steel high-pressure reactor.
9786
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9785 – 9788

Nickel-Catalysed Hydroalkoxylation of 1,3-Butadiene
COMMUNICATION
varied with respect to the nickel loading. The amount of
sodium borohydride needed is clearly catalytic and the best
results were obtained with a NaBH4/Ni ratio ranging from
0.5 to 1. A higher ratio led to drastically reduced conver-
sions that can be attributed to nickel boride formation. Nev-
ertheless, high selectivities for butenyl ethers were main-
tained. The best ligand/nickel ratio was found to be around
1.5 equivalents and larger amounts did not improve selectiv-
ity and were shown to be detrimental to 1,3-butadiene con-
version. Finally, the amount of 1,3-butadiene was increased
to 400 equivalents with respect to nickel (Table 2, entries 6
and 7) which required twice as much ethanol to limit the
formation of dimers and to obtain almost unchanged selec-
tivities compared to the previous tests with 80 equivalents of
1,3-butadiene.
cohols, as exemplified by phenol, failed to yield the desired
products.
Hydroalkoxylation of butadiene occurs with high selectivi-
ty for the formation of butenyl ethers. The nickel precursor
and ligand used are readily available and are easily reacted
with sodium borohydride to generate the catalyst. A strong
ligand dependency has been observed and the use of the
dppb ligand is critical to access highly selective catalysts
under relatively mild reaction conditions. Further work to
gain a better understanding of the specific role of the ligand
in the unprecedented selectivity revealed, as well as further
improvements, are underway.
Experimental Section
We further studied the reaction with other alcohols
(Table 3). Methanol failed to react if the catalyst was pre-
pared in this solvent. Methanol appears to react with
In a typical catalytic experiment, the nickel precursor NiACHTNUTRGNEUNG(acac)2 (35 mg,
0.14 mmol), the diphosphine (0.21 mmol) and sodium borohydride
(NaBH4; 5.3 mg; 0.14 mmol), were introduced into a Schlenk tube closed
with a rotaflo stop cock. Distilled and degassed ethanol (10 mL), previ-
ously cooled to 08C, was then added. The tube was further cooled down
to À208C. A precise volume of butadiene (11.1 mmol, 0.6 g, 1 mL) was
condensed in a Schlenk tube cooled with an acetone/dry ice mixture and
transferred into the reaction mixture. Finally the reactor was heated at
808C and stirred with a magnetic stirrer for 17 h. After the reaction, the
system was cooled and excess gaseous butadiene was vented before chro-
matographic analysis.
Table 3. Hydroalkoxylation reaction of various alcohols with the Ni-
ACHTUNGTRENNUNG
(acac)2/dppb/NaBH4 catalytic system.[a]
Entry
Alcohol/co-solvent
Butadiene
Conv. [%]
4+5
[%]
OC8
[%]
C8
[%]
1
2
3
4
MeOH
0
89
95
0
15
0
0
98
98
0
75
0
0
<2
<2
0
25
0
0
<2
<2
0
<2
0
MeOH/THF[b]
PhCH2OH
iPrOH
5
iPrOH/EtOH[c]
tBuOH
6[b]
7[c]
Acknowledgements
PhOH/THF[b]
0
0
0
0
We are grateful to the Region Nord Pas de Calais for S. Bigot and A. Mi-
fleur’s fellowship, as well as the CNRS (S.B.). We acknowledge the Minis-
try of Research and Technology, the CNRS and the Institut Universitaire
de France for their financial support.
[a] Conditions: butadiene=11.1 mmol, Ni/butadiene=1:80, alcohol=
10 mL, T=808C, t=17 h. [b] The nickel precursor was formed in 2 mL
THF at room temperature for 10 min, and ethanol (10 mL) or phenol
(5 g) were added. [c] The reaction was run with iPrOH (5 mL) and EtOH
(5 mL), only ethoxylation products were obtained.
Keywords: butadiene · ethers · homogeneous catalysis ·
hydroalkoxylation · nickel
sodium borohydride before the nickel(II) precursor is re-
duced as evidenced by the rapid dihydrogen evolution that
was observed in the early seconds of reaction. The catalyst
was therefore generated beforehand by the reaction of
[1] a) A. Behr, M. Becker, T. Beckmann, L. Johnen, J. Lechinski, S.
Reyer, Angew. Chem. 2009, 121, 3652; Angew. Chem. Int. Ed. 2009,
48, 3598; for pioneering work, see: b) S. Takahashi, Tetrahedron
Lett. 1967, 8, 2451.
NaBH4 and NiACHTUNGTRENNUNG(acac)2 in THF. This alternative procedure al-
lowed the ready synthesis of methyl butenyl ethers with
high selectivities and yields that are comparable to those ob-
tained with ethanol. From our results, primary alcohols react
readily in the hydroalkoxylation reaction and benzyl alcohol
also afforded a mixture of butenyl ethers in good yields. In
contrast, more highly substituted alcohols failed to react and
no reaction products (butenyl ethers, dimerization or telo-
merisation products) were formed with isopropanol and tert-
butanol and prior preparation of the catalyst in an aprotic
solvent did not result in any improvement. A reaction in-
volving a 1:1 mixture of ethanol and isopropanol was run to
gain more insight into the nature of the reaction. Only the
products of ethoxylation (4 and 5) were observed, in low
yield, along with the OC8 telomerisation products, showing
that isopropanol is not only unreactive but is also detrimen-
tal to the catalyst activity and selectivity. Lastly, aromatic al-
[2] a) J.-L. Tschan, E. J. Garcia-Suarez, Z. Freixa, H. Launay, H.
Hagen, J. Benet-Buchholz, P. W. N. M. van Leeuwen, J. Am. Chem.
Soc. 2010, 132, 6463; b) F. Benvenuti, C. Carlini, M. Lami, M.
Marchionna, R. Patrini, A. M. Raspolli Galletti, G. Sbrana, J. Mol.
Catal. A Chem. 1999, 144, 27.
[3] a) N. D. Clement, L. Routaboul, A. Grotevendt, R. Jackstell, M.
Beller, Chem. Eur. J. 2008, 14, 7408; b) J. Mesnager, P. Lammel, E.
Jeanneau, C. Pinel, Appl. Catal. A 2009, 368, 22; c) A. Grotevendt,
R. Jackstell, D. Michalik, M. Gomez, M. Beller, ChemSusChem
2009, 2, 63.
[4] a) P. C. A. Bruijnincx, R. Jastrzebski, P. J. C. Hausoul, R. J. M. Klein
Gebbink, B. M. Weckhuysen in Topics in Organometallic Chemistry,
Vol. 39: Organometallics and Renewables (Eds.: M. A. R. Meier,
B. M. Weckhuysen, P. C. A. Bruijnincx), Springer, Berlin, 2012,
pp. 45; b) I. Pennequin, J. Meyer, I. Suisse, A. Mortreux, J. Mol.
Catal. A Chem. 1997, 120, 139; c) S. Bigot, J. Lai, M. Sauthier, A.
Mortreux, Y. Castanet, Appl. Catal. A 2010, 382, 181.
[5] a) P. Denis, A. Mortreux, F. Petit, G. Buono, G. Peiffer, J. Org.
Chem. 1984, 49, 5274; b) M. A. El Amrani, A. Mortreux, F. Petit,
Chem. Eur. J. 2013, 19, 9785 – 9788
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9787

I. Suisse, M. Sauthier et al.
Tetrahedron Lett. 1989, 30, 6515; c) P. Denis, A. Jean, J. F. Croizy, A.
Mortreux, F. Petit, J. Am. Chem. Soc. 1990, 112, 1292; d) M. A.
El Amrani, I. Suisse, N. Knouzi, A. Mortreux, Tetrahedron Lett.
1995, 36, 5011.
[13] V. J. Beger, C. Duschek, H. Fꢃllbier, W. Gaube, J. Prakt. Chem.
(Leipzig) 1974, 316, 1, 26.
[14] I. Suisse, H. Bricout, A. Mortreux, Tetrahedron Lett. 1994, 35, 413.
[15] M. Beller, A. Krotz, W. Baumann, Adv. Synth. Catal. 2002, 344, 517.
[16] a) Y. Kawabata, T. Hayashi, J. Ogata, J. Chem. Soc. Chem.
Commun. 1979, 462; b) Y. Kawabata, T. Hayashi, T. Isoyama, J.
Ogata, Bull. Chem. Soc. Jpn. 1981, 54, 3438.
[17] E. Paumard, A. Mortreux, F. Petit, J. Chem. Soc. Chem. Commun.
1989, 1380.
[18] G. P. C. M. Dekker, C. J. Cornelis, K. Vrieze, P. W. N. M. Van Leeu-
wen, Organometallics 1992, 11, 1598.
[6] A. Mortreux in Chemistry for the 21st Century: Transition Metal Cat-
alyzed Reactions (Eds: S.-I. Murahashi, S. G. Davies), Blackwell,
Oxford, 1999, pp. 159–169.
[7] K. C. Dewhirst, J. Org. Chem. 1967, 32, 1297.
[8] E. J. Smutny, J. Am. Chem. Soc. 1967, 89, 6793.
[9] R. Patrini, M. Lami, M. Marchionna, F. Benvenuti, A. M. Galletti,
G. Sbrana, J. Mol. Catal. A Chem. 1998, 129, 179.
[10] T. C. Shields, W. E. Walker, J. Chem. Soc. D 1971, 193.
[11] F. J. Weigert, W. C. Drinkard, J. Org. Chem. 1973, 38, 335.
[12] D. Commereuc, Y. Chauvin, Bull. Soc. Chim. Fr. 1974, 3–4, 652.
[19] B. Ganem, J. O. Osby, Chem. Rev. 1988, 88, 763.
Received: February 18, 2013
Published online: June 12, 2013
9788
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9785 – 9788